SOLUTION:
SOLUTION:
PC had worked fine to years then got these symptoms:
Tried to swap memory stick from another from identical PC: no help.
Download Hardware Maintenance and User Guides - 2 beeps code not in there !
SOLUTION: new BIOS battery + reset BIOS and all OK now.
First create transparent image to overlay using Gimp (Colors / Colors to Alpha, then export as GIF)
Then open PDF in LibreOffice Writer, Left-click on page / Edit with External Tool to open Gimp (if this does not work, then see below):
Switch back to PDF in LibreOffice Writer:
Check new PDF in your favourite PDF viewer.
My first try was this, but it did not work for me:
> xdg-mime query filetype <x.jpg>
> xdg-mime query default image/jpeg
> xdg-mime default /usr/share/applications/gimp.desktop image/jpeg
> xdg-mime query default image/jpeg
> xdg-open <x.jpg> # does not use gimp :-(
7z e -oMyDir X.exejava -Xmx512M -jar mkgmap.jar --family-id=<nnnn> --family-name=MyOSM --gmapsupp 6*.img clas*.TYP where <nnnn> is the first 4 digits of the nnnn...img files(Probably works for other devices.)
You can "test" the map first by using Qlandkarte: File / Load Map / navigate to your newly created map directory / mapsetc.tdb / mapsetc.img / Open
"<<ERROR>> Grab failed in function" <your function> "unable to execute immediate action"
SOLUTION: To continue working with this session: restart FVWM, either from mouse, or remotely login, set DISPLAY, kill fvwm and start it again (in background). Or switch to another console. Probably best to do a clean login when you can !
SOLUTION: reformat card: in my case I used gParted under Linux (also worked for a 64GB card when formatted to FAT32).
[fvwm][__execute_function]: <<ERROR>> No such command 'WindowListFunc'Problem: it seems the FVWM defaults path is hard coded at build time. If this path does not work then this message is seen.
SOLUTION: add this line to the start of your fvwm2rc:
Read <insert your path here>/share/fvwm/ConfigFvwmDefaultsYou probably also want to set your modules path too:
ModulePath <insert your path here>/libexec/fvwm/2.6.2
usb 1-2: can't set config #1, error -110Spent a lot of time Googling, lots of advice tried: upgraded udev and kernel, checked correct modules loaded and kernel options selected. Some people had success with different USB cables or rebooting PC, but for me:
SOLUTION: Disconnect from PC, reboot IdeaPad and it just worked !!
Unrecognized file type?
Problem: by default gpsbabel downloads TCX files as type "Course".
SOLUTION: Convert existing tracks into "Activity" and upload again:
> gpsbabel -i gtrnctr -f myFile-crs.tcx -o gtrnctr,course=no -F myFile-act.tcx
SOLUTION pt1 Libreoffice 3.5 has moved it's Python script location to $HOME/.config/libreoffice/3/user/Scripts/python so your scripts go in here.
SOLUTION pt2 Python 2.7.x needed (for my Libreoffice 3.5 at least). I had 2.5, 2.6 and 3.2 installed. I added 2.7.3 (latest in Gentoo 2.7.x and my Python scripts appeared after restarting Libreoffice.
To confirm the Python version:
> strace -Ff -o t.t libreoffice > egrep 'python|ENOENT' t.t
SOLUTION Edit / Preferences / Appearence / Text widget options add "-font {Courier -10}" (without the quotes) Or "-*-lucidatypewriter-medium-r-*-*-10-*-*-*-*-*-*-*" (with quotes).
firefox -remote 'OpenUrl(about:blank)' Error: No running window foundI read lots of bug reports/fixes but discovered this was my problem (obvious when you see it, although I have had it set since 2006 and it work fine till recently):
SOLUTION ensure that you have not set MOZ_NO_REMOTE in the environment where you start Firefox (local, global, or in a script).
(Comment on Mozilla bug 733565 suggests this originates from Firefox 10 (which sounds about right to me, from memory of when it stopped working for me)).
style "user-font" {
font_name = " Sans 8"
}
widget_class "*" style "user-font"
gtk-font-name="Sans 8"
From ISO, created bootable SD card with usb-creator-gtk. Live OS boots OK, but installation fails with:
- GUI reports: "ubi-partman failed"
- /var/log/syslog has the message: "no space left on device"
- df -h : shows both root filesystem 100%
SOLUTION When creating bootable card use option "Stored in extra reserved space": I select 550MB, but less would probably do.
The listing order seems to be the same as ls -Ul (-U is directory order, unsorted)
SOLUTION various scripts can be found to use copy or move to re-order the files, or try FATsort (in Gentoo unmask sys-fs/fatsort, emerge and run, for example, /usr/sbin/fatsort /dev/sdc1 as root on an unmounted drive).
This is not a bootable disk. Please insert a bootable floppy and press any key to try again.
SOLUTION Instead of using my Gentoo/AMD64 box to create the SD card setup I used another PC with Ubuntu/x86 8.10, reformated the (cheapy) SD card to FAT32 (using gparted) and reinstalled the ISO, again using UNetbootin. It now booted fine now on the eeePC 4G/701, and also installed fine to the internal drive.
Why ? I have no idea: either different versions of some software or else UNetbootin creates some kind of AMD, or 64-bit, incompatibility.
> firefox -remote 'openURL(index.html, new-tab)' Error: Failed to send command: 500 command not parseable
SOLUTION (not ideal !):
> firefox -remote "openURL(/home/simonk/index.html, new-tab)" > firefox -remote "openURL($PWD/index.html, new-tab)"
:-( /dev/dvdrw: media is not recognized as recordable DVD: 0Tried lots of things suggested on the web: different disks, burning programs, wiggling wires around, even bought a new burner, but in the end:
SOLUTION: wait a minute or so after inserting the blank DVD and closing the DVD tray before trying to burn !!
...
File "//usr/lib64/python2.5/site-packages/pytrainer/lib/xmlUtils.py", line 127, in shortFromGPS
tree = xml.etree.cElementTree.parse(gtrnctrFile).getroot()
File "<string>", line 45, in parse
File "<string>", line 32, in parse
SyntaxError: unbound prefix: line 13, column 8
SOLUTION: in /usr/share/pytrainer/plugins/garmin-hr/main.py change join(lines[6:]) to join(lines[2:])
...
File "//usr/lib64/python2.5/site-packages/pytrainer/lib/xmlUtils.py", line 127, in shortFromGPS
tree = xml.etree.cElementTree.parse(gtrnctrFile).getroot()
File "<string>", line 45, in parse
File "<string>", line 32, in parse
SyntaxError: unbound prefix: line 17, column 8
SOLUTION: in /usr/share/pytrainer/plugins/garmin-hr/main.py update line to: (single line)
<TrainingCenterDatabase xmlns="http://www8.garmin.com/xmlschemas/TrainingCenterDatabasev2" xmlns:xsi="http://www.w3.org/2001/XMLSchema-instance" xsi:schemaLocation="http://www8.garmin.com/xmlschemas/TrainingCenterDatabasev2 http://www8.garmin.com/xmlschemas/TrainingCenterDatabasev2.xsd">
PROBLEM: still no data in app
pytrainer.sh -d => log.out
... 2009-06-04 22:23:54,548|INFO|record|importFromGTRNCTR|Retrieving data from /tmp/file_mod.gtrnctr 2009-06-04 22:23:58,223|DEBUG|xmlUtils|shortFromGPS|>> 2009-06-04 22:23:58,224|DEBUG|xmlUtils|shortFromGPS|parsing /tmp/file_mod.gtrnctr 2009-06-04 22:23:58,224|DEBUG|xmlUtils|shortFromGPS|Retrieving sport info 2009-06-04 22:23:58,453|DEBUG|xmlUtils|shortFromGPS|Retrieved info: [] 2009-06-04 22:23:58,453|DEBUG|xmlUtils|shortFromGPS|<< 2009-06-04 22:23:58,469|INFO|record|importFromGTRNCTR|GPS: 0 entries found 2009-06-04 22:23:58,470|INFO|record|importFromGTRNCTR|No tracks found in GPS device 2009-06-04 22:23:58,470|DEBUG|record|importFromGTRNCTR|<<CAUSE: GPSbabel 1.3.6 changed format of TrainingCenterDatabase to v2 of the Garmin XML. Pytrainer needs updating to read this new format.
SOLUTION: I have not tried it, but reverting to GPSBabel 1.3.5 should work till Pytrainer is updated
program error: "KeyError: 'prf_us_system'"SOLUTION: Manually edit conf.xml and add:
prf_us_system="True"
"Warning: disk_free_space() [function.disk-free-space]: Permission denied..."
SOLUTION: Ensure directory DIR_DOWNLOAD in /var/www/wtorrent/user.conf.php is readable to the webserver user (e.g. www-data).
(This is answered in several bug reports, but a Google of the above message does not seem to find them !!)
Caused by Adblock Filterset.G Updater Add-On for Firefox trying to update. www.pierceive.com.nyud.net is the project homepage www.pierceive.com accessed via Coral Cache. See also this thread on the Filterset.G forum.
I *think* it can be safely ignored !! (but compare the filterset version reported by Firefox compared with that on the website). (If Firefox hangs then update to the latest version of Adblock Filterset.G Updater.)
Dec 22 15:28:58 zoxed sg_write: data in/out 56/56 bytes for SCSI command 0x12--guessing data in; Dec 22 15:28:58 zoxed program cdparanoia not setting count and/or reply_len properly Dec 22 15:28:58 zoxed sg_write: data in/out 26/26 bytes for SCSI command 0x5a--guessing data in; Dec 22 15:28:58 zoxed program cdparanoia not setting count and/or reply_len properly ... Dec 22 15:29:03 zoxed sg_write: data in/out 30576/30576 bytes for SCSI command 0xbe--guessing data in; Dec 22 15:29:03 zoxed program cdparanoia not setting count and/or reply_len properly Dec 22 15:29:08 zoxed printk: 186 messages suppressed. ...Kernel 2.6.19, pure SATA setup. /dev/cdrom links to /dev/sr0. Same problem if /dev/sg1 is used.
Grip (2.5.1) uses: cdparanoia -d /dev/sg1 13:[.0]-13:[.17041] X.wav -Z ? add -g, no /dev/scdN
SOLUTION: (Dec 2007) Upgrade to Kernel 2.6.22-gentoo-r8 did not help, but upgrading cdparanoia from 3.9.8-r5 to 3.10-pre0 worked.
SOLUTION: check if another process has taken hold of the device by running "fuser /dev/dvd". In my case, after my burning finished Grip had polled the device and taken it. I stopped Grip and the eject worked. Update: eject /dev/dvd still failed, but eject /dev/cdrom works. Both are softlinks to the same device /dev/sr0 !!
SOLUTION: using the GRUB device instead of the OS device did work, eg: "grub-install fd0".
SOLUTION: the onboard sound for the nVidia MCP51 is in fact supported by hda-intel. Later kernels (eg 2.6.25) need CONFIG_SND_HDA_INTEL and CONFIG_SND_HDA_CODEC_ANALOG.
SOLUTION: hack the WizzRSS CSS theme file:
If this does not work:
SOLUTION: I had recently added export
GREP_OPTIONS="--color=always" to /etc/profile. But the outputting
of the escape codes for colour quietly breaks one of the kernel build
scripts. Remove GREP_OPTIONS and re-run make
modules_install and the modules should now be generated.
(The grep that fails is in line 12 of
/usr/src/linux/scripts/Makefile.modinst - you can confirm it by
checking the value of __modules from make -p
modules_install)
hdparm /dev/hdc gave "HDIO_GETGEO failed: Inappropriate ioctl for device". Googling around suggested a readahead problem, but "hdparm -A0 /dev/hdc" also fails !!
System log gave things like:
hdc: command error: status=0x51 { DriveReady SeekComplete Error }
hdc: command error: error=0x54 { AbortedCommand LastFailedSense=0x05 }
ide: failed opcode was: unknown
end_request: I/O error, dev hdc, sector 0
Buffer I/O error on device hdc, logical block 0
Upgrading cdrecord, growisofs, the kernel and hdparm did not help !
SOLUTION: use "regionset" to set the DVD drives region (by
default none was set) (the tool says you *MUST* have a DVD in the
drive when you run it) and disable DMA. Works OK so far to burn a DVD.
(although hdparm still gives the same error as above)
(With DMA enabled System log gives this when I try growisofs:
hdc: DMA timeout retry
hdc: timeout waiting for DMA
hdc: status timeout: status=0xd0 { Busy }
ide: failed opcode was: unknown
hdc: drive not ready for command
hdc: ATAPI reset complete
Update: with DMA enabled mount a CD works, eventually by the driver turing off the DMA ("hdc: DMA disabled"). Mounting or buring a DVD with DMA enabled gives "ide-cd: cmd 0x28 timed out" and "hdc: DMA timeout retry". With DMA disabled it works OK, but slowly (DVD burning at x1.4).
"Cannot read partial active file from nntp server"And my NNTP Groups where all empty.
SOLUTION: Turned out to be an problem with my NNTP provider: my account had been temporarily disabled, so the password authentication failed. Not obvious from what the message said, but it is correct !!
Jan 21 15:56:03 localhost hdc: tray open Jan 21 15:56:03 localhost end_request: I/O error, dev hdc, sector 64 Jan 21 15:56:03 localhost isofs_fill_super: bread failed, dev=hdc, iso_blknum=16, block=16Googling "tray open" etc revealed very little. Installed a CDROM drive from another machine and that worked OK. Checking Kernel code (linux/drivers/ide/ide-cd.c) did not give any ideas.
SOLUTION: replaced the drive with a DVD burner (Samsung) and it worked OK (this Philips drive was very loud anyway !!). (But see above re-"regionset".)
menupopup {
font-family: fixed !important;
font-size: 16 !important
}
menubar, menubutton, menulist, menu, menuitem {
font-family: fixed !important;
font-size: 15 !important;
}
toolbar {
font-family: fixed !important;
font-size: 14 !important;
}
.tab-text {
font-family: fixed !important;
font-size: 14 !important;
}
(I got the information from unpacking an existing theme: unzip the
.jar file, collect all the .css files and grep around till you find
what you need, then stick in userChrome.css)
My solution: get a *real* Window Manager (FVWM) :-) and hack around with something like the following (to get the window name to help create the regexp use xwininfo):
DestroyFunc MaxFullScreen
AddToFunc MaxFullScreen
+ I WindowStyle NoTitle, BorderWidth 0
+ I Maximize true
+ I ResizeMove 100 100 0 0
DestroyFunc FileSelectorFunct
AddToFunc FileSelectorFunct
+ I ResizeMove 730p 745p -0p -0p
DestroyFunc FuncFocusWindow
AddToFunc FuncFocusWindow
+ I ThisWindow ("*Firefox") Function MaxFullScreen
+ I ThisWindow ("Open*") Function FileSelectorFunct
Style * OverrideGrabFocus
AddToFunc MaxFullScreen + I WindowStyle NoTitle, BorderWidth 0, HandleWidth 0 + I UpdateStyles # resize with new borders etc before doing Max + I Maximize true + I ResizeMove 100 100 0 0 AddToFunc UnFullScreen + I WindowStyle Title, BorderWidth 3, HandleWidth 3 + I ResizeMove 50 50 25 25 # set start size + I Resize # now user resizes window Key f A SC Pick FullScreen Key u A SC Pick UnFullScreen
Works fine using USB module usb-storage etc (Kernel 2.6.10) (but I have only tested CF and SD sockets) (check you also have CONFIG_SCSI_MULTI_LUN enabled).
I later noticed that the actual card device (/dev/sdx1) does not appear directly: a scan needs to be provocated. How ?
Check what the permissions/group are for /dev/sd*. Mine are 660/disk (from udev?) so I had to add myself to the "disk" group.
(16-in-1, 3.5 inch, floppy bay size, can be used as external, USB 2.0 connection. ("Vendor=0dda ProdID=2027"))
xorg.conf:
Section "InputDevice"
Identifier "MouseOpti"
Driver "mouse"
Option "Protocol" "auto"
Option "ZAxisMapping" "4 5"
Option "Device" "/dev/input/usbRecv"
EndSection
udev rule:
BUS="usb", SYSFS{product}="USB RECEIVER", NAME="%k", SYMLINK="input/usbRecv"
TAO (track at once) worked fine. (Looks like either a cdrecord bug, or a hardware problem.)
Also resmgrd was not running (this may be my fault from messing with rc to speed up the start up). The effect was that it looked like the CD was being burnt, but it never stopped. And kill -9 was no use: needed a reboot. And the disk was a coaster.
SUSE 9.1 Professional, k3b v?, cdrecord "Cdrecord-Clone-dvd 2.01a27 (i686-suse-linux)".
Note: Gentoo / cdrecord "Cdrecord-Clone 2.01" reports "Supported modes: TAO PACKET SAO SAO/R96P SAO/R96R" i.e. NOT DAO.
I guess you can see how to configure it !!
(defun my-check-FCC () "maybe alter default FCC depending on To: address" (interactive) (cond ((string-match (concat (regexp-quote "myName@") "\\|" (regexp-quote "myOtherName@") "\\|" (regexp-quote "commonName@yahoo.co.uk")) (vm-mail-mode-get-header-contents "To:")) (beginning-of-buffer) (if (re-search-forward "FCC: .*$") (replace-match (concat "FCC: " vm-folder-directory "mine") nil nil)) ))) (add-hook 'mail-send-hook 'my-check-FCC)
Also be aware that when Kinternet does not show connected when you think it should it is because PPPoE it is setup to only connect on demand (I wasted a lot of time here trying to tget it to show connected :-(
(Basic Eur30 card: AGP-8, 64Mb, DVI output (DVI-D), CRT output (D-Sub), and TV out (Composite).)
Whilst migrating from lilo to grub I found this problem and solution: no amount of googling helped me :-(
My setup:
SOLUTION: Enable BIOS to recognise the drive that contains the kernel. Now SUSE boots OK.
My h/w setup:
Video Card: Elsa Winner 1000 T3D / 4mb SDRAM / S3 Trio3D/2x chipset
Monitor: Belinea 103010 / 17inch, max 1280x1024
My s/w setup:
Xfree86 4.3.0, driver: s3virge
SOLUTION: the only solutions I have found is to either run at 1024x768, or to set Option "ShadowFB" "on" in XF86Config (this then allows up to 1280x1024 to be used).
December 2004 update: I upgraded to Xorg: no change (I suspect the driver has been abandoned now :-).
materm: can't open pseudo-tty
strace showed it trying to open /dev/ptmx but this did not exist for me. As far as I can see this is trying to access a "Unix98" pseudo terminal and requires the kernel to be built with CONFIG_UNIX98_PTYS and CONFIG_DEVPTS_FS (see aterm Unix98 PTY README).
SOLUTION: Looking for a quicker fix I checked the rxvt code that I have and this lead me to a quicker solution: edit src/command.c and allow getpt() to be used (replace #ifdef HAVE_GETPT with #if 1. Worked for me. (A more correct fix would be to update the configure script to check for this and set HAVE_GETPT correctly (it does in rxvt))).
Also note that multi-aterm is very fussy about the case of the
Xdefaults: including the True/False (to be fair I now notice that rxvt
also is, but the two are not compatible (they use the same name, but
different case !!). But you can use, for example,
materm.scrollbar: False).
But so far it works great :-) although the man page is a little sparce (eg how to set font size, tab title etc).
alias sound-slot-0 ymfpci (ymfpci being the kernel module for my device)I think it works like this: